Industrial equipment cooperative control method and system based on low codes

Abstract

The invention relates to the technical field of equipment cooperative control, and discloses an industrial equipment cooperative control method and system based on low codes. The method comprises the steps that triggering condition parameters, optimization target parameters and constraint boundary parameters are extracted from an intention description template of a low-code platform to construct an intention triple, the intention triple and capacity description vectors of all industrial devices are input into a constraint satisfaction solver to be solved, a task allocation matrix and a target parameter value sequence are obtained, and a target parameter value sequence is obtained; and according to the real-time operation parameters and the real-time position coordinates of each industrial device, constructing a global collaborative state matrix, comparing the global collaborative state matrix with the difference of the task allocation matrix to obtain an increment adjustment instruction set, querying a protocol adaptation matrix, translating the increment adjustment instruction set into a control message, and issuing the control message. According to the invention, the automatic construction capability, the real-time adaptive scheduling capability and the unified control capability of cross-protocol equipment of the industrial equipment cooperative control system are improved.

Description

Technical Field

[0001] This application relates to the field of equipment collaborative control technology, and in particular to a low-code-based method and system for collaborative control of industrial equipment. Background Technology

[0002] Collaborative control of industrial equipment is a core component of modern intelligent manufacturing. In existing technologies, the construction of collaborative control systems for industrial equipment mainly relies on professional developers to write customized control code for specific production line scenarios. Engineers need to master the communication protocols, control parameter ranges, and linkage logic of each piece of equipment on the production line in advance, and manually configure the trigger rules and execution order between the equipment one by one. Some systems have introduced low-code platforms to replace some of the code writing work with a graphical drag-and-drop method. Engineers complete the arrangement of collaborative control logic by connecting equipment nodes and filling in trigger condition parameters in a visual interface. The above methods provide basic tool support for the construction of collaborative control systems for industrial equipment.

[0003] However, existing technologies have the following shortcomings: Existing low-code platforms require engineers to know the complete device collaboration logic in advance before orchestration. As the number of production line devices increases, the orchestration complexity increases exponentially. Engineers still need to manually configure each trigger rule and each device node one by one. Essentially, it only replaces writing code with drawing flowcharts, without fundamentally reducing the threshold for building collaborative control logic. In addition, the collaborative control logic orchestrated by existing low-code platforms is stored in the form of static flowcharts, which cannot perceive the real-time status changes and dynamic position offsets of each industrial device during operation. When the device's operating parameters or spatial position change, it cannot adaptively adjust the task allocation scheme. More importantly, industrial devices of different brands on the production line use different communication protocols. Existing platforms need to configure communication parameters separately for each protocol device. The adaptation work for multi-protocol heterogeneous devices severely restricts the cross-device reuse capability of collaborative control logic. Summary of the Invention

[0004] This application provides a low-code-based collaborative control method and system for industrial equipment, which solves the problems in existing low-code platforms such as the reliance on manual static orchestration by engineers for collaborative control logic, the inability to dynamically adjust task allocation based on real-time equipment status, and the complexity of adapting to heterogeneous devices with multiple protocols. It improves the automated construction capability, real-time adaptive scheduling capability, and unified control capability of collaborative control systems for industrial equipment across protocols.

[0005] Firstly, this application provides a low-code-based collaborative control method for industrial equipment, the low-code-based collaborative control method for industrial equipment comprising: Step S1: Extract trigger condition parameters, optimization target parameters, and constraint boundary parameters from the intent description template of the low-code platform to construct intent triples; Step S2: Input the intent triplet and the capability description vector of each industrial device in the device registry into the constraint satisfaction solver. Solve the problem with the optimization target parameter as the objective function, the constraint boundary parameter as the constraint condition, and the adjustable parameter value of each device as the decision variable. Automatically translate the obtained target parameter value sequence and task allocation relationship into node connection relationship and trigger condition configuration in the low code orchestration diagram to obtain the task allocation matrix and target parameter value sequence. Step S3: Construct a global collaborative state matrix based on the real-time operating parameters and real-time position coordinates of each industrial device, and compare the differences between the global collaborative state matrix and the task allocation matrix to obtain an incremental adjustment instruction set; Step S4: Query the protocol adaptation matrix according to the incremental adjustment instruction set, translate the incremental adjustment instruction set into a control message, and then send it out.

[0006] Secondly, this application provides a low-code-based industrial equipment collaborative control system, which includes: The extraction module is used to extract trigger condition parameters, optimization target parameters, and constraint boundary parameters from the intent description template of the low-code platform to construct intent triples; The input module is used to input the intent triplet and the capability description vector of each industrial device in the device registry into the constraint satisfaction solver. The solver uses the optimization target parameter as the objective function, the constraint boundary parameter as the constraint condition, and the adjustable parameter value of each device as the decision variable. The solved target parameter value sequence and task allocation relationship are automatically translated into node connection relationship and trigger condition configuration in the low code orchestration diagram to obtain the task allocation matrix and target parameter value sequence. The comparison module is used to construct a global collaborative state matrix based on the real-time operating parameters and real-time location coordinates of each industrial device, and compare the global collaborative state matrix with the task allocation matrix to obtain an incremental adjustment instruction set. The translation module is used to query the protocol adaptation matrix according to the incremental adjustment instruction set, translate the incremental adjustment instruction set into control messages, and then send them out.

[0007] The technical solution provided in this application upgrades the engineer's operational objects in the low-code platform from manually orchestrated control logic to structured intent triplets. The engineer's control objectives are encapsulated in three structures: trigger condition parameters, optimization target parameters, and constraint boundary parameters. The driving constraint satisfaction solver automatically derives the equipment collaborative control logic using adjustable parameters of each device as decision variables, optimization target parameters as the objective function, and constraint boundary parameters as constraints. The solution results are automatically translated into node connections and trigger condition configurations in the low-code orchestration diagram. This fundamentally eliminates the prerequisite that engineers must know the complete collaborative logic in advance to complete the orchestration, solving the scalability bottleneck of existing low-code platforms where orchestration complexity increases exponentially with the number of devices. Furthermore, by collecting real-time operating parameters and real-time position coordinates of each industrial device at a fixed sampling period to construct a global collaborative state matrix, and incorporating real-time position coordinates as an independent feature dimension into the matrix, a difference comparison with the task allocation matrix is ​​performed to generate an incremental adjustment instruction set. This enables collaborative control decisions to perceive the dynamic changes in the spatial position of the devices in real time, overcoming the inherent defect of existing low-code platforms' static orchestration logic in being unable to respond to real-time changes in the operating state of devices.

[0008] The constraint satisfaction solver performs optimization on the objective function within a legal search space defined by constraint boundary parameters. This transforms the engineer's control objectives, expressed in natural language, into a sequence of target parameter values ​​for each device and a task allocation matrix. This enables automatic derivation from control intent to executable collaborative logic, a feature that eliminates reliance on the engineer's personal experience in generating collaborative control logic. Simultaneously, the protocol adaptation matrix uses abstract control command types as rows and industrial communication protocol types as columns. It encodes the parameter adjustment values ​​of each device in the incremental adjustment command set into control messages of the corresponding protocol format according to the conversion rules stored in the corresponding matrix elements. These messages are then sequentially distributed after being arranged in ascending order of the device's baseline response delay time. This achieves unified translation and distribution of a single collaborative control logic across multiple brands of heterogeneous devices. Furthermore, it inherently supports an online hot update mechanism that generates incremental messages only for devices that have undergone changes, allowing adjustments to the collaborative control logic to be made without downtime and redeployment. Attached Figure Description

[0009] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0010] Figure 1 This is a schematic diagram of one embodiment of the low-code-based collaborative control method for industrial equipment in this application. Figure 2This is a schematic diagram of the normalization result of the global cooperative state matrix in an embodiment of this application. Detailed Implementation

[0011] This application provides a low-code-based collaborative control method and system for industrial equipment. The terms "first," "second," "third," "fourth," etc. (if present) in the specification, claims, and accompanying drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments described herein can be implemented in a sequence other than that illustrated or described herein. Furthermore, the terms "comprising" or "having" and any variations thereof are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or device that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or devices.

[0012] For ease of understanding, the specific process of the embodiments of this application is described below. Please refer to [link / reference]. Figure 1 One embodiment of the low-code-based collaborative control method for industrial equipment in this application includes: Step S1: Extract trigger condition parameters, optimization target parameters, and constraint boundary parameters from the intent description template of the low-code platform to construct intent triples; Specifically, the intent description template of the low-code platform is a standardized front-end interface for engineers to input control objectives. The template includes a device identifier input area, a monitoring parameter type selection area, and a trigger threshold input area. The content entered by the engineer in the template is parsed and extracted into three structures: trigger condition parameters, optimization target parameters, and constraint boundary parameters. These three are combined and encapsulated to form an intent triple. The core function of the intent triple is to transform the control objectives expressed by the engineer in natural language into structured inputs that satisfy the constraints and can be computed by the solver. The trigger condition parameters define the initiation timing of the solution, the optimization target parameters define the objective function of the solution, and the constraint boundary parameters define the legal value range of the decision variables.

[0013] Step S2: Input the intent triplet and the capability description vector of each industrial device in the device registry into the constraint satisfaction solver. Solve the problem with the optimization target parameter as the objective function, the constraint boundary parameter as the constraint condition, and the adjustable parameter value of each device as the decision variable. The solved target parameter value sequence and task allocation relationship are automatically translated into node connection relationship and trigger condition configuration in the low code orchestration diagram to obtain the task allocation matrix and target parameter value sequence. Specifically, the constraint satisfaction solver takes the intent triplet and the capability description vector of each industrial device as input, and the adjustable parameter values ​​of each device as decision variables. Within the range of values ​​defined by the constraint boundary parameters, it solves for the parameter combination that optimizes the objective function defined by the objective parameters. The solution results include two types of outputs: first, a sequence of objective parameter values ​​for each device arranged by device number; and second, a task allocation matrix encoded with the task allocation relationships between devices. Both types of outputs are simultaneously translated into node connection relationships and trigger condition configurations in a low-code orchestration diagram, presented to engineers for confirmation in a visual format.

[0014] Step S3: Construct a global collaborative state matrix based on the real-time operating parameters and real-time position coordinates of each industrial device, compare the differences between the global collaborative state matrix and the task allocation matrix, and obtain the incremental adjustment instruction set. Specifically, the global collaborative state matrix collects current, speed, temperature, load rate, and real-time position coordinates of each industrial device at a fixed sampling period. Each parameter is normalized according to its historical minimum and maximum values. The normalized state feature vectors of all devices are then concatenated into a matrix in device number order. The real-time position coordinates of each device in the global collaborative state matrix are extracted, and the Euclidean distance between each device and the task target position is calculated. The updated Euclidean distance replaces the original distance parameter in the task allocation matrix, and the matrix is ​​recalculated. The old and new task allocation matrices are compared element-by-element, and incremental adjustment instructions are generated only for devices that have changed. These instructions are then encapsulated into a set of incremental adjustment instructions.

[0015] Step S4: Query the protocol adaptation matrix according to the incremental adjustment instruction set, translate the incremental adjustment instruction set into control messages and send them out.

[0016] Specifically, the protocol adaptation matrix uses abstract control command types as rows and industrial communication protocol types as columns. Matrix elements store corresponding message conversion rules, which include the target register address, data type, byte order, and checksum calculation method. Based on the communication protocol type and command type of each device in the incremental adjustment command set, matrix elements are located. The incremental parameter adjustment values ​​are encoded into binary control messages according to the corresponding conversion rules, and then sequentially sent out in ascending order of each device's baseline response delay time to ensure timing alignment of the coordinated actions of all devices.

[0017] In one specific embodiment, step S1 includes: Identify the device identifier, monitoring parameter type, and trigger threshold from the intent description template, and encode the device identifier, monitoring parameter type, and trigger threshold into a trigger condition structure to obtain the trigger condition parameters; Based on the device identifier in the trigger condition parameters, query the device registry, extract the rated operating parameter range of the corresponding industrial equipment, and encode the upper limit of rated current, upper limit of rated speed and upper limit of rated temperature of each industrial equipment into a constraint boundary structure to obtain the constraint boundary parameters. Identify the optimization index type and preservation coefficient from the intent description template, encode the optimization index type and preservation coefficient into an optimization target structure, and obtain the optimization target parameters; The trigger condition parameters, constraint boundary parameters, and optimization objective parameters are combined and encapsulated to obtain the intent triplet.

[0018] Specifically, the intent description template is a standardized form in the low-code platform's front-end interface for engineers to fill in control objectives. The template includes three fields: a device identifier field, a monitoring parameter type dropdown selection field, and a trigger threshold value input field. The platform reads the device identifier, monitoring parameter type, and trigger threshold entered by the engineer from these three fields, respectively, and encodes them into a trigger condition structure according to a predefined structure format. The trigger condition structure has the following fields: device identifier field, monitoring parameter type field, and trigger threshold field. These three fields together constitute the trigger condition parameters, which explicitly define when which parameter of which device reaches which value to initiate the subsequent collaborative control process.

[0019] Based on the device identifier field in the trigger condition parameters, the registration record of the corresponding industrial equipment is located in the device registry. From this registration record, three rated operating parameters—rated current upper limit, rated speed upper limit, and rated temperature upper limit—are extracted. These three parameters are then encoded into a constraint boundary structure according to a predefined structure format. The fields of the constraint boundary structure correspond sequentially to the rated current upper limit, rated speed upper limit, and rated temperature upper limit fields, respectively. These three fields together constitute the constraint boundary parameters. The values ​​of the constraint boundary parameters are derived from the factory rated parameters written into the registration of each industrial device in the device registry, rather than being manually entered by engineers, thus eliminating the risk of errors introduced by manual entry.

[0020] The optimization index type and maintenance coefficient filled in by the engineer are read from the optimization target input area of ​​the intent description template. The optimization index type represents the category of production index to be optimized in this collaborative control, and the maintenance coefficient represents the minimum proportion value that the index must maintain after the trigger condition is met. The optimization index type and maintenance coefficient are encoded into an optimization target structure according to a predefined structure format, forming the optimization target parameters. The three structures of trigger condition parameters, constraint boundary parameters, and optimization target parameters are combined and encapsulated according to the predefined format of intent triples. The trigger condition parameters, as the first element, define the start time; the optimization target parameters, as the second element, define the solution objective; and the constraint boundary parameters, as the third element, define the legal range of the decision variables. The three together constitute the intent triple.

[0021] In one specific embodiment, step S2 involves inputting the intent triplet and the capability description vectors of each industrial device in the device registry into the constraint satisfaction solver, including: Extract the device type, adjustable parameter range, communication protocol type, and baseline response delay time of each industrial device from the device registry, and concatenate the device type, adjustable parameter range, communication protocol type, and baseline response delay time into a capability description vector; The optimization objective parameters in the intention triplet are transformed into objective functions with the adjustable parameter values ​​of each device as decision variables, and the constraint boundary parameters are transformed into value constraints for each decision variable. The objective function, value constraints, and capability description vector are input into the constraint satisfaction solver to obtain the target parameter values ​​and task allocation relationships for each industrial device.

[0022] Specifically, the registration record for each industrial device in the device registry includes four fields: device type, adjustable parameter range, communication protocol type, and baseline response delay time. The device type field records the functional role of the device in the warehousing production line; the adjustable parameter range field records the types of parameters that the device is allowed to adjust under normal operating conditions, along with their upper and lower limits; the communication protocol type field records the industrial protocol standard used for communication between the device and the low-code platform; and the baseline response delay time field records the baseline time interval (in milliseconds) from receiving a control message to completing an action. The values ​​of these four fields are concatenated in a fixed order: device type, adjustable parameter range, communication protocol type, and baseline response delay time, forming the device's capability description vector. The dimensions of the capability description vector correspond one-to-one with the number of fields, and the values ​​for each dimension are directly derived from the registration record in the device registry without any transformation.

[0023] The optimization index type field and the maintenance coefficient field of the objective parameters in the intention triple are transformed into the basis for constructing the objective function. The objective function uses the adjustable parameter values ​​of each industrial equipment as decision variables, the production index calculation formula corresponding to the optimization index type as the function expression, and the product of the maintenance coefficient and the current measured index value as the lower bound constraint of the objective function, forming an optimization objective function with lower bound constraints. The rated current upper limit field, rated speed upper limit field, and rated temperature upper limit field of the constraint boundary parameters in the intention triple are transformed into upper bound constraints for the corresponding decision variables. The value range of each decision variable is provided by the adjustable parameter range field in the capability description vector as the lower bound and by the constraint boundary parameters as the upper bound. Together, they constitute the complete value constraints for each decision variable.

[0024] The constraint-satisfaction solver receives three types of inputs: an objective function, value constraints, and a capability description vector. The value constraints define the legal search space for each decision variable. Within this legal search space, the solver performs optimization on the objective function and outputs a combination of target parameter values ​​for each industrial device that satisfies all value constraints and maximizes the objective function. Simultaneously, based on the device type and adjustable parameter range fields in the capability description vector of each industrial device, the solver determines the specific responsibilities of each device in this collaborative control task and encodes these responsibilities into task allocation relationships.

[0025] In one specific embodiment, step S2 involves automatically translating the solved target parameter value sequence and task allocation relationship into node connection relationships and trigger condition configurations in a low-code orchestration graph, including: Based on the collaborative execution order among the industrial devices in the task allocation relationship, the node connection relationship between the corresponding device nodes is generated in the low-code orchestration diagram. Based on the trigger condition parameters in the intent triple, write the trigger condition configuration in the property panel of the corresponding device node in the low-code orchestration diagram; The task allocation relationship is encoded into a task allocation matrix, and the target parameter values ​​of each industrial device are arranged in order of device number to obtain a sequence of target parameter values.

[0026] Specifically, the task allocation relationship records the division of responsibilities and execution order of each industrial device in this collaborative control task. The low-code orchestration diagram expresses the collaborative execution relationship between devices in the form of a directed graph. Each node in the graph corresponds to one industrial device, and the directed edges between nodes correspond to the collaborative execution order between devices. Based on the execution order of each industrial device in the task allocation relationship, a directed edge is generated between two adjacent device nodes in the low-code orchestration diagram. The starting point of the directed edge corresponds to the device node that is executed first, and the ending point corresponds to the device node that is executed later. All device nodes and the directed edges between them together constitute the node connection relationship. The node connection relationship is presented visually on the canvas of the low-code orchestration diagram. Engineers complete the orchestration by confirming the operation, without the need for manual connection.

[0027] The trigger condition parameters in the intent triple include three fields: device identifier, monitoring parameter type, and trigger threshold. Locate the device node corresponding to the device identifier field in the low-code orchestration diagram, open the property panel of that device node, write the value of the monitoring parameter type field to the monitoring parameter configuration item in the property panel, and write the value of the trigger threshold field to the trigger threshold configuration item in the property panel. After writing, the trigger condition configuration of that device node will be consistent with the trigger condition parameters in the intent triple. During operation, the platform will monitor the monitoring parameters of the industrial equipment corresponding to that device node in real time. When the monitoring parameter value meets the trigger threshold, the subsequent collaborative control process will be automatically started.

[0028] The task allocation relationship is encoded in the form of a two-dimensional matrix. The row index of the matrix corresponds to the industrial equipment number, and the column index corresponds to the task target number. The matrix elements take values ​​of 0 or 1. When the i-th industrial equipment is assigned to the s-th task target, the element in the i-th row and s-th column of the matrix has a value of 1; otherwise, it has a value of 0. After encoding, the task allocation matrix is ​​obtained. The target parameter values ​​of each industrial equipment come from the solution output of the constraint satisfaction solver. According to the ascending order of the equipment number in the equipment registry, the target parameter values ​​of each equipment are arranged sequentially to form a target parameter value sequence. The i-th element of the target parameter value sequence corresponds to the target parameter value of the industrial equipment with equipment number i in this collaborative control task.

[0029] In one specific embodiment, after obtaining the task allocation matrix and the target parameter value sequence in step S2, the method further includes: When the optimization objective parameters in the intent description template change, the objective function is reconstructed based on the changed optimization objective parameters. The reconstructed objective function and the constraint boundary parameters are then input into the constraint satisfaction solver for re-solution, resulting in an updated objective parameter value sequence and an updated task assignment matrix. The updated task allocation matrix is ​​compared element by element with the task allocation matrix to extract the set of devices whose task allocation relationships have changed. The incremental parameter adjustment value of each device is calculated based on the difference between the updated target parameter value sequence and the target parameter value sequence. The incremental parameter adjustment values ​​of each device in the device set are encapsulated into incremental adjustment instructions, which are then sorted in ascending order according to the baseline response delay time and updated to the corresponding node connection relationship and trigger condition configuration in the low-code orchestration diagram, resulting in the updated task allocation matrix and target parameter value sequence.

[0030] Specifically, when an engineer modifies the optimization index type field or the preservation coefficient field of the optimization objective parameter in the intent description template on the low-code platform's front-end interface, the platform detects the change in the optimization objective parameter. It then reconstructs the objective function using the changed optimization index type and preservation coefficient. The difference between the reconstructed objective function and the original objective function lies only in the change of the preservation coefficient value or the optimization index type; the rest of the objective function structure remains unchanged. The reconstructed objective function and the original constraint boundary parameters are input into the constraint satisfaction solver. The constraint boundary parameters remain unchanged in this re-solution because the rated operating parameter range of the industrial equipment has not changed. The solver performs optimization on the reconstructed objective function within the search space defined by the original value constraints, outputting an updated sequence of objective parameter values ​​and an updated task assignment matrix.

[0031] The updated task allocation matrix is ​​compared element-wise with the task allocation matrix obtained in step S2. Each element in the same row and column is checked for consistency. If the element value in the i-th row and s-th column of the updated task allocation matrix differs from that in the task allocation matrix, the device number i is added to the set of devices whose task allocation relationship has changed. This process is repeated for all rows and columns to obtain the complete set of devices. For each device in the set, the updated target parameter value at the corresponding position is extracted from the updated target parameter value sequence, and the original target parameter value at the corresponding position is extracted from the target parameter value sequence. The difference between the two values ​​is calculated to obtain the incremental parameter adjustment value for that device. The sign of the incremental parameter adjustment value indicates the direction of parameter adjustment, and the absolute value indicates the magnitude of parameter adjustment.

[0032] Each device in the device set has its device identifier and corresponding incremental parameter adjustment value encapsulated into an incremental adjustment instruction according to a predefined instruction format. The incremental adjustment instruction contains two fields: a device identifier field and an incremental parameter adjustment value field. All incremental adjustment instructions from the device set are sorted in ascending order based on the baseline response delay time recorded in the capability description vector for each device, with devices having smaller baseline response delay times listed later and those with larger baseline response delay times listed earlier. Incremental adjustment instructions are issued sequentially in this order to ensure that devices with different baseline response delay times can complete parameter adjustments within a similar timeframe after receiving the instruction. After issuance, the updated task allocation matrix replaces the stored task allocation matrix, the updated target parameter value sequence replaces the stored target parameter value sequence, and the node connection relationships and trigger condition configurations between changed device nodes in the low-code orchestration diagram are synchronously updated to the changed content.

[0033] In one specific embodiment, step S3 involves constructing a global collaborative state matrix based on the real-time operating parameters and real-time position coordinates of each industrial device, including: The system collects current, speed, temperature, load rate, and real-time position coordinates from each industrial device at a fixed sampling period, and then concatenates these data into a state feature vector for each industrial device. Based on the historical minimum and maximum values ​​of each parameter in the state feature vector of each industrial equipment, interval normalization is performed on each parameter value in the state feature vector to obtain the normalized state feature vector. The normalized state feature vectors of all industrial equipment are concatenated in order of equipment number to obtain the global cooperative state matrix. The real-time position coordinates of each industrial device are extracted from the global collaborative state matrix. The Euclidean distance between each industrial device and each task target position is calculated. The distance parameters in the task allocation matrix are updated with the Euclidean distance and then re-input into the constraint satisfaction solver for solving. The updated task allocation matrix obtained by resolving is compared element by element with the task allocation matrix to extract the device number and corresponding parameter adjustment value of the changed task allocation relationship, and encapsulate them into an incremental adjustment instruction set.

[0034] Specifically, five types of parameters—current, rotational speed, temperature, load rate, and real-time position coordinates—are read from the data acquisition interface of each industrial device at a fixed sampling period (unit: milliseconds). The real-time position coordinates include both the horizontal and vertical coordinates of the device within the warehouse coordinate system. These five types of parameters are then concatenated in a fixed order: current, rotational speed, temperature, load rate, horizontal coordinate, and vertical coordinate, forming a state feature vector for the industrial device at the current sampling moment. This state feature vector has a fixed six-dimensional dimension, with each dimension corresponding one-to-one with the corresponding parameter type.

[0035] For each dimension of the state feature vector, interval normalization is performed on the parameter values. The normalization process uses the minimum value of that dimension parameter in historical operating data as the lower bound and the maximum value as the upper bound. The normalized result for that dimension parameter is obtained by subtracting the historical minimum value from the current parameter value and dividing by the difference between the historical maximum and minimum values. The normalized result is a real number between 0 and 1. This normalization calculation is performed sequentially on the parameter values ​​of all six dimensions of the state feature vector to obtain the normalized state feature vector of the industrial equipment. The normalized state feature vectors of all industrial equipment on the production line are then concatenated into rows of a matrix in ascending order of equipment number. The number of rows in the matrix equals the total number of industrial equipment, and the number of columns is fixed at six, resulting in the global collaborative state matrix.

[0036] Extract the values ​​from the fifth and sixth columns of the corresponding row of each industrial device from the global collaborative state matrix to obtain the real-time x-coordinate and y-coordinate values ​​of each industrial device. For each industrial device and each task target location, calculate the Euclidean distance between them. The Euclidean distance is calculated by adding the square of the difference between the industrial device's x-coordinate value and the task target's x-coordinate value, and the square root of the difference between the industrial device's y-coordinate value and the task target's y-coordinate value. Replace the original distance parameters in the task assignment matrix with the recalculated Euclidean distances between all industrial devices and all task target locations. Re-input the updated task assignment matrix into the constraint satisfaction solver to obtain the updated task assignment matrix. The updated task allocation matrix is ​​compared element by element with the original task allocation matrix. The device number corresponding to the row index where the element value has changed is extracted. The parameter values ​​of the corresponding devices are extracted from the updated target parameter value sequence and the original target parameter value sequence, and the difference is calculated to obtain the incremental parameter adjustment value of each device. Each device number and the corresponding incremental parameter adjustment value are encapsulated into an incremental adjustment instruction. The set of all incremental adjustment instructions is the incremental adjustment instruction set.

[0037] For example, at the end of the previous sampling period, an AGV was assigned to shelf number 3 to perform a storage and retrieval task. The horizontal and vertical coordinate values ​​of its state feature vector were normalized and written into the corresponding row of the global cooperative state matrix. At the end of the current sampling period, the real-time position coordinates of the AGV were extracted from the global cooperative state matrix. It was found that its position had moved from the previous period toward shelf number 3. At the same time, the real-time position coordinates of another AGV showed that its Euclidean distance to shelf number 3 was less than the Euclidean distance between the aforementioned AGV and shelf number 3. After the constraint satisfaction solver re-solved based on the updated Euclidean distance, the storage and retrieval task of shelf number 3 was reassigned to the closer AGV. The corresponding element values ​​in the updated task assignment matrix changed. The platform extracted the incremental parameter adjustment values ​​of the two AGVs and encapsulated them into an incremental adjustment instruction set.

[0038] Figure 2 This is a schematic diagram illustrating the normalization result of the global cooperative state matrix in an embodiment of this application. For example... Figure 2As shown, the normalization result of the global collaborative state matrix constructed for 6 AGVs in this embodiment of the application is illustrated. The rows of the matrix correspond to 6 industrial devices, AGV-01 to AGV-06, and the columns correspond to 6 state feature dimensions, namely current value, speed value, temperature value, load rate, horizontal axis, and vertical axis. The value of each element in the matrix is ​​the result of interval normalization processing of the parameter value of the corresponding feature dimension of the corresponding device according to the historical minimum and historical maximum values ​​of each parameter. The value range is a real number between 0 and 1. The darker the color, the larger the normalized value, and the lighter the color, the smaller the normalized value. In the figure, the normalized value of the vertical axis of AGV-02 is 0.97 and the normalized value of the current value of AGV-03 is 0.83, indicating that the real-time parameter values ​​of the above devices in the corresponding feature dimensions are close to the historical maximum values. Based on this, the platform identifies significant differences in device status and triggers the subsequent difference comparison process of the task allocation matrix.

[0039] In one specific embodiment, step S4, querying the protocol adaptation matrix based on the incremental adjustment instruction set, includes: A protocol adaptation matrix is ​​pre-constructed. The rows of the protocol adaptation matrix correspond to the abstract control instruction type, and the columns of the protocol adaptation matrix correspond to the industrial communication protocol type. The matrix elements of the protocol adaptation matrix store the conversion rules for translating the corresponding abstract control instruction type into the corresponding industrial communication protocol type message format. The conversion rules include the target register address, data type, byte order, and checksum calculation method. The device registry is queried based on the device identifier of each device in the incremental adjustment instruction set to obtain the communication protocol type of each device. The corresponding abstract control instruction type is determined based on the instruction content of each instruction in the incremental adjustment instruction set. The protocol adaptation matrix is ​​queried based on the communication protocol type and the abstract control instruction type to obtain the corresponding conversion rules. The parameter adjustment values ​​of each instruction in the incremental adjustment instruction set are encoded into control messages in the corresponding industrial communication protocol format according to the conversion rules, and then the control messages are sent out in ascending order according to the reference response delay time of each device.

[0040] Specifically, the protocol adaptation matrix is ​​pre-built in the form of a two-dimensional matrix and stored in the configuration database of the low-code platform. The row index of the matrix corresponds to the number of the abstract control instruction type, which includes four types: speed adjustment instruction, target position instruction, start / stop instruction, and access trigger instruction, with row indices from 1 to 4. The column index of the matrix corresponds to the number of the industrial communication protocol type, which includes four types: proprietary TCP protocol brand A, proprietary TCP protocol brand B, MQTT protocol, and Modbus-TCP protocol, with column indices from 1 to 4. The matrix elements store the conversion rules for the corresponding row and column positions. The conversion rules consist of four fields: the target register address field records the starting address of the register where data is written when executing this type of instruction under this protocol type; the data type field records the encoding format of the parameter adjustment value in the message (INT16, INT32, or FLOAT32); the byte order field records the arrangement order of multi-byte data in the message (big-endian or little-endian); and the checksum calculation method field records the checksum algorithm type used for message integrity verification (CRC16 or CRC32).

[0041] Based on the device identifier field of each incremental adjustment instruction in the incremental adjustment instruction set, the registration record of the corresponding industrial equipment is located in the device registry. The industrial communication protocol type used by the device is read from the communication protocol type field of the registration record, obtaining the column index corresponding to the device. The instruction content of the incremental adjustment instruction is used to determine which of the four abstract control instruction types it belongs to: speed adjustment instruction, target position instruction, start / stop instruction, and access trigger instruction, obtaining the corresponding row index. The corresponding matrix element in the protocol adaptation matrix is ​​located using the row and column indices, and the conversion rules stored in that matrix element are read. The four field values ​​of the conversion rules serve as the basis for subsequent control message encoding.

[0042] The parameter adjustment value of each instruction in the incremental adjustment instruction set is converted into a binary value according to the encoding format specified by the data type field of the corresponding conversion rule. The binary value is then rearranged byte by byte according to the byte order specified by the byte order field. The rearranged binary value is written to the register address offset specified by the target register address field. The checksum is calculated on the complete message content according to the checksum calculation method field and appended to the end of the message to obtain a control message conforming to the corresponding industrial communication protocol format. The control messages corresponding to all devices in the incremental adjustment instruction set are sorted in descending order according to the baseline response delay time recorded by each device in the capability description vector. The control message corresponding to the device with the largest baseline response delay time is sent first, and the control message corresponding to the device with the smallest baseline response delay time is sent last. The control messages are sent to each industrial device in this order through the communication interface of the corresponding industrial communication protocol to ensure that industrial devices with different baseline response delay times can complete the parameter adjustment action at similar times after receiving the control message.

[0043] The above describes the low-code-based collaborative control method for industrial equipment in the embodiments of this application. The following describes the low-code-based collaborative control system for industrial equipment in the embodiments of this application. One embodiment of the low-code-based collaborative control system for industrial equipment in the embodiments of this application includes: The extraction module is used to extract trigger condition parameters, optimization target parameters, and constraint boundary parameters from the intent description template of the low-code platform to construct intent triples; The input module is used to input the intent triplet and the capability description vector of each industrial device in the device registry into the constraint satisfaction solver. The solver uses the optimization target parameter as the objective function, the constraint boundary parameter as the constraint condition, and the adjustable parameter value of each device as the decision variable. The solved target parameter value sequence and task allocation relationship are automatically translated into node connection relationship and trigger condition configuration in the low code orchestration diagram to obtain the task allocation matrix and target parameter value sequence. The comparison module is used to construct a global collaborative state matrix based on the real-time operating parameters and real-time location coordinates of each industrial device, and compare the global collaborative state matrix with the task allocation matrix to obtain an incremental adjustment instruction set. The translation module is used to query the protocol adaptation matrix according to the incremental adjustment instruction set, translate the incremental adjustment instruction set into control messages, and then send them out.

[0044] In one specific embodiment, trigger condition parameters, optimization target parameters, and constraint boundary parameters are extracted from the intent description template of the low-code platform to construct an intent triple, including: The device identifier, monitoring parameter type, and trigger threshold are identified from the intent description template, and the device identifier, monitoring parameter type, and trigger threshold are encoded into a trigger condition structure to obtain the trigger condition parameter; Based on the device identifier in the trigger condition parameters, query the device registry, extract the rated operating parameter range of the corresponding industrial equipment, and encode the upper limit of rated current, upper limit of rated speed and upper limit of rated temperature of each industrial equipment into a constraint boundary structure to obtain the constraint boundary parameters. The optimization index type and the retention coefficient are identified from the intent description template, and the optimization index type and the retention coefficient are encoded into an optimization target structure to obtain the optimization target parameters. The trigger condition parameters, the constraint boundary parameters, and the optimization target parameters are combined and encapsulated to obtain the intent triplet.

[0045] In one specific embodiment, the intent triple and the capability description vectors of each industrial device in the device registry are input into a constraint satisfaction solver, including: Extract the device type, adjustable parameter range, communication protocol type, and baseline response delay time of each industrial device from the device registry, and concatenate the device type, adjustable parameter range, communication protocol type, and baseline response delay time into the capability description vector; The optimization objective parameters in the intent triplet are transformed into objective functions with the adjustable parameter values ​​of each device as decision variables, and the constraint boundary parameters are transformed into value constraints for each decision variable. The objective function, the value constraints, and the capability description vector are input into the constraint satisfaction solver for solving, thereby obtaining the target parameter values ​​and task allocation relationships for each industrial device.

[0046] The above embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A low-code-based collaborative control method for industrial equipment, characterized in that, The method includes: Step S1: Extract trigger condition parameters, optimization target parameters, and constraint boundary parameters from the intent description template of the low-code platform to construct intent triples; Step S2: Input the intent triplet and the capability description vector of each industrial device in the device registry into the constraint satisfaction solver. Solve the problem with the optimization target parameter as the objective function, the constraint boundary parameter as the constraint condition, and the adjustable parameter value of each device as the decision variable. Automatically translate the obtained target parameter value sequence and task allocation relationship into node connection relationship and trigger condition configuration in the low code orchestration diagram to obtain the task allocation matrix and target parameter value sequence. Step S3: Construct a global collaborative state matrix based on the real-time operating parameters and real-time position coordinates of each industrial device, and compare the differences between the global collaborative state matrix and the task allocation matrix to obtain an incremental adjustment instruction set; Step S4: Query the protocol adaptation matrix according to the incremental adjustment instruction set, translate the incremental adjustment instruction set into a control message, and then send it out.

2. The low-code-based collaborative control method for industrial equipment according to claim 1, characterized in that, Step S1 includes: The device identifier, monitoring parameter type, and trigger threshold are identified from the intent description template, and the device identifier, monitoring parameter type, and trigger threshold are encoded into a trigger condition structure to obtain the trigger condition parameter; Based on the device identifier in the trigger condition parameters, query the device registry, extract the rated operating parameter range of the corresponding industrial equipment, and encode the upper limit of rated current, upper limit of rated speed and upper limit of rated temperature of each industrial equipment into a constraint boundary structure to obtain the constraint boundary parameters. The optimization index type and the retention coefficient are identified from the intent description template, and the optimization index type and the retention coefficient are encoded into an optimization target structure to obtain the optimization target parameters. The trigger condition parameters, the constraint boundary parameters, and the optimization target parameters are combined and encapsulated to obtain the intent triplet.

3. The low-code-based collaborative control method for industrial equipment according to claim 1, characterized in that, In step S2, the intent triple and the capability description vectors of each industrial device in the device registry are input into the constraint satisfaction solver, including: Extract the device type, adjustable parameter range, communication protocol type, and baseline response delay time of each industrial device from the device registry, and concatenate the device type, adjustable parameter range, communication protocol type, and baseline response delay time into the capability description vector; The optimization objective parameters in the intent triplet are transformed into objective functions with the adjustable parameter values ​​of each device as decision variables, and the constraint boundary parameters are transformed into value constraints for each decision variable. The objective function, the value constraints, and the capability description vector are input into the constraint satisfaction solver for solving, thereby obtaining the target parameter values ​​and task allocation relationships for each industrial device.

4. The low-code-based collaborative control method for industrial equipment according to claim 3, characterized in that, In step S2, the obtained sequence of target parameter values ​​and task allocation relationships are automatically translated into node connection relationships and trigger condition configurations in the low-code orchestration graph, including: Based on the collaborative execution order among the industrial devices in the task allocation relationship, the node connection relationship between the corresponding device nodes is generated in the low-code orchestration diagram. Based on the trigger condition parameters in the intent triple, write the trigger condition configuration into the property panel of the corresponding device node in the low-code orchestration diagram; The task allocation relationship is encoded into the task allocation matrix, and the target parameter values ​​of each industrial device are arranged in order of device number to obtain the target parameter value sequence.

5. The low-code-based collaborative control method for industrial equipment according to claim 4, characterized in that, In step S2, after obtaining the task allocation matrix and the target parameter value sequence, the method further includes: When the optimization objective parameters in the intent description template change, the objective function is reconstructed based on the changed optimization objective parameters. The reconstructed objective function and the constraint boundary parameters are then input into the constraint satisfaction solver for re-solving to obtain the updated objective parameter value sequence and the updated task allocation matrix. The updated task allocation matrix is ​​compared element by element with the task allocation matrix to extract the set of devices whose task allocation relationships have changed. The incremental parameter adjustment value of each device is calculated based on the difference between the updated target parameter value sequence and the target parameter value sequence. The incremental parameter adjustment values ​​of each device in the device set are encapsulated into incremental adjustment instructions, which are then sorted in ascending order according to the baseline response delay time and updated to the corresponding node connection relationship and trigger condition configuration of the low-code orchestration graph, resulting in the updated task allocation matrix and the target parameter value sequence.

6. The low-code-based collaborative control method for industrial equipment according to claim 5, characterized in that, In step S3, the global collaborative state matrix is ​​constructed based on the real-time operating parameters and real-time position coordinates of each industrial device, including: Current, rotational speed, temperature, load rate, and real-time position coordinates are collected from each industrial device at a fixed sampling period. The current, rotational speed, temperature, load rate, and real-time position coordinates are then concatenated to form a state feature vector for each industrial device. Based on the historical minimum and maximum values ​​of each parameter in the state feature vector of each industrial equipment, interval normalization is performed on each parameter value in the state feature vector to obtain a normalized state feature vector. The normalized state feature vectors of all industrial equipment are concatenated in order of equipment number to obtain the global cooperative state matrix; The real-time position coordinates of each industrial device are extracted from the global collaborative state matrix. The Euclidean distance between each industrial device and each task target position is calculated. The distance parameters in the task allocation matrix are updated with the Euclidean distance and then re-input into the constraint satisfaction solver for solving. The updated task allocation matrix obtained by resolving is compared element by element with the task allocation matrix. The device number and corresponding parameter adjustment value of the changed task allocation relationship are extracted and encapsulated into the incremental adjustment instruction set.

7. The low-code-based collaborative control method for industrial equipment according to claim 6, characterized in that, In step S4, querying the protocol adaptation matrix based on the incremental adjustment instruction set includes: The protocol adaptation matrix is ​​pre-constructed, where the rows of the protocol adaptation matrix correspond to abstract control instruction types, the columns of the protocol adaptation matrix correspond to industrial communication protocol types, and the matrix elements of the protocol adaptation matrix store conversion rules for translating the corresponding abstract control instruction type into the corresponding industrial communication protocol type message format. The conversion rules include target register address, data type, byte order, and checksum calculation method. The device registry is queried based on the device identifier of each device in the incremental adjustment instruction set to obtain the communication protocol type of each device. The corresponding abstract control instruction type is determined based on the instruction content of each instruction in the incremental adjustment instruction set. The protocol adaptation matrix is ​​queried based on the communication protocol type and the abstract control instruction type to obtain the corresponding conversion rule. The parameter adjustment values ​​of each instruction in the incremental adjustment instruction set are encoded into control messages in the corresponding industrial communication protocol format according to the conversion rules, and the control messages are sent out sequentially after being arranged in ascending order according to the reference response delay time of each device.

8. A low-code-based collaborative control system for industrial equipment, characterized in that, For implementing the low-code-based collaborative control method for industrial equipment as described in any one of claims 1-7, the low-code-based collaborative control system for industrial equipment comprises: The extraction module is used to extract trigger condition parameters, optimization target parameters, and constraint boundary parameters from the intent description template of the low-code platform to construct intent triples; The input module is used to input the intent triplet and the capability description vector of each industrial device in the device registry into the constraint satisfaction solver. The solver uses the optimization target parameter as the objective function, the constraint boundary parameter as the constraint condition, and the adjustable parameter value of each device as the decision variable. The solved target parameter value sequence and task allocation relationship are automatically translated into node connection relationship and trigger condition configuration in the low code orchestration diagram to obtain the task allocation matrix and target parameter value sequence. The comparison module is used to construct a global collaborative state matrix based on the real-time operating parameters and real-time location coordinates of each industrial device, and compare the differences between the global collaborative state matrix and the task allocation matrix to obtain an incremental adjustment instruction set. The translation module is used to query the protocol adaptation matrix according to the incremental adjustment instruction set, translate the incremental adjustment instruction set into control messages, and then send them out.

9. The system according to claim 8, characterized in that, Extract trigger condition parameters, optimization target parameters, and constraint boundary parameters from the intent description template of the low-code platform to construct an intent triple, including: The device identifier, monitoring parameter type, and trigger threshold are identified from the intent description template, and the device identifier, monitoring parameter type, and trigger threshold are encoded into a trigger condition structure to obtain the trigger condition parameter; Based on the device identifier in the trigger condition parameters, query the device registry, extract the rated operating parameter range of the corresponding industrial equipment, and encode the upper limit of rated current, upper limit of rated speed and upper limit of rated temperature of each industrial equipment into a constraint boundary structure to obtain the constraint boundary parameters. The optimization index type and the retention coefficient are identified from the intent description template, and the optimization index type and the retention coefficient are encoded into an optimization target structure to obtain the optimization target parameters. The trigger condition parameters, the constraint boundary parameters, and the optimization target parameters are combined and encapsulated to obtain the intent triplet.

10. The system according to claim 8, characterized in that, The intent triple and the capability description vectors of each industrial device in the device registry are input into the constraint satisfaction solver, including: Extract the device type, adjustable parameter range, communication protocol type, and baseline response delay time of each industrial device from the device registry, and concatenate the device type, adjustable parameter range, communication protocol type, and baseline response delay time into the capability description vector; The optimization objective parameters in the intent triplet are transformed into objective functions with the adjustable parameter values ​​of each device as decision variables, and the constraint boundary parameters are transformed into value constraints for each decision variable. The objective function, the value constraints, and the capability description vector are input into the constraint satisfaction solver for solving, thereby obtaining the target parameter values ​​and task allocation relationships for each industrial device.

Images